Structural and catalytic characterization of pichia stipitis OYE 2.6, a useful biocatalyst for asymmetric alkene reductions

Article abstract of DOI:10.1002/adsc.201200213

We have probed Pichia stipitis CBS 6054 Old Yellow Enzyme 2.6 (OYE 2.6) by several strategies including X-ray crystallography, ligand binding and catalytic assays using the wild-type as well as libraries of site-saturation mutants. The alkene reductase crystallized in space group P 6₃ 2 2 with unit cell dimensions of 127.1×123.4 A and its structure was solved to 1.5 A resolution by molecular replacement. The protein environment surrounding the flavin mononucleotide (FMN) cofactor was very similar to those of other OYE superfamily members; however, differences in the putative substrate binding site were also observed. Substrate analog complexes were analyzed by both UV-Vis titration and X-ray crystallography to provide information on possible substrate binding interactions. In addition, four active site residues were targeted for site saturation mutagenesis (Thr 35, Ile 113, His 188, His 191) and each library was tested against three representative Baylis-Hillman adducts. Thr 35 could be replaced by Ser with no change in activity; other amino acids (Ala, Cys, Leu, Met, Gln and Val) resulted in diminished catalytic efficiency. The Ile 113 replacement library yielded a range of catalytic activities, but had very little impact on stereoselectivity. Finally, the two His residues (188 and 191) were essentially intolerant of substitutions with the exception of the His 191 Asn mutant, which did show significant catalytic ability. Structural comparisons between OYE 2.6 and Saccharomyces pastorianus OYE1 suggest that the key interactions between the substrate hydroxymethyl groups and the side-chain of Thr 35 and/or Tyr 78 play an important role in making OYE 2.6 an (S)-selective alkene reductase. Copyright

Full text of DOI:10.1002/adsc.201200213

FULL PAPERS
DOI: 10.1002/adsc.201200213
Structural and Catalytic Characterization of Pichia stipitis OYE
2.6, a Useful Biocatalyst for Asymmetric Alkene Reductions
a,b
a,b
a,b
a,
Yuri A. Pompeu, Bradford Sullivan, Adam Z. Walton, and Jon D. Stewart *
1
a
26 Sisler Hall, Department of Chemistry, University of Florida, Gainesville, FL 32611, USA
Fax : (+1)-352-846-0743; phone: (+1)-352-846-0743; e-mail: jds2@chem.ufl.edu
These authors contributed equally to this work
b
Received: March 14, 2012; Revised: May 17, 2012; Published online: && &&, 0000
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201200213.
Abstract: We have probed Pichia stipitis CBS 6054 Baylis–Hillman adducts. Thr 35 could be replaced by
Old Yellow Enzyme 2.6 (OYE 2.6) by several strat- Ser with no change in activity; other amino acids
egies including X-ray crystallography, ligand binding (Ala, Cys, Leu, Met, Gln and Val) resulted in dimin-
and catalytic assays using the wild-type as well as li- ished catalytic efficiency. The Ile 113 replacement li-
braries of site-saturation mutants. The alkene reduc- brary yielded a range of catalytic activities, but had
tase crystallized in space group P6 22 with unit cell very little impact on stereoselectivity. Finally, the
3
dimensions of 127.1ꢀ123.4 ꢁ and its structure was two His residues (188 and 191) were essentially intol-
solved to 1.5 ꢁ resolution by molecular replacement. erant of substitutions with the exception of the His
The protein environment surrounding the flavin mo- 191 Asn mutant, which did show significant catalytic
nonucleotide (FMN) cofactor was very similar to ability. Structural comparisons between OYE 2.6 and
those of other OYE superfamily members; however, Saccharomyces pastorianus OYE1 suggest that the
differences in the putative substrate binding site key interactions between the substrate hydroxymeth-
were also observed. Substrate analog complexes yl groups and the side-chain of Thr 35 and/or Tyr 78
were analyzed by both UV-Vis titration and X-ray play an important role in making OYE 2.6 an (S)-se-
crystallography to provide information on possible lective alkene reductase.
substrate binding interactions. In addition, four
active site residues were targeted for site saturation
mutagenesis (Thr 35, Ile 113, His 188, His 191) and Keywords: alkenes; biocatalysis; enzyme catalysis;
each library was tested against three representative reduction; structural biology; X-ray diffraction
Introduction
a collection of native alkene reductases and applying
protein engineering strategies to existing enzymes.
An ever-increasing number of putative old yellow
Alkene reductase enzymes have recently emerged as
powerful catalysts for asymmetric C=C reductions enzyme superfamily members is available from
[
1]
(
for recent reviews, see ref. ). Pioneering work on genome sequencing projects. We and others have
[
2]
Saccharomyces pastorianus
Old Yellow Enzyme cloned, overexpressed and characterized these pro-
1
(OYE1) by Massey established that this flavoprotein teins with respect to their substrate and stereoselec-
reduced a variety of electron-deficient alkenes and tivities in the hopes of identifying pairs of enantio-
[6]
also catalyzed a mechanistically related dismutation complementary biocatalysts (summarized in ref. ). In
reaction in the absence of the physiological reductant the early stages, target genes were chosen on the basis
[3]
NADPH. Since then, several groups have explored of DNA availability and similarity to S. pastorianus
synthetic applications of S. pastorianus OYE1 (for OYE1. While some stereochemical diversity was ob-
[
4]
recent examples, see ref. and references cited there- served, the vast majority of old yellow enzyme homo-
in). While a number of successes have been reported, logs had essentially the same catalytic properties.
a single biocatalyst can only provide one product Some of our earliest protein engineering studies re-
[5]
enantiomer. We have explored two potential strat- vealed that replacing an active site Trp residue (posi-
egies for overcoming this limitation: accumulating tion 116) with Ile radically changed the stereoselectiv-
[
7]
ity of S. pastorianus OYE1, an observation later
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
ÞÞ
These are not the final page numbers!

Yuri A. Pompeu et al.
FULL PAPERS
Table 1. OYE 2.6 data collection and refinement statistics.
Accession code
3TJL
3UPW
4DF2
Observed active site ligand
X-ray source
malonate
NSLS X25
nicotinamide/malonate
NSLS X25
p-chlorophenol
Rigaku CuKa
Space group
P6 22
P6 22
P6 22
3
3
3
Unit cell dimensions
a=b, c (ꢁ)
127.16, 122.45
44.10–1.50
92,358 (2,365)
99.07 (96.0)
7.6 (4.7)
0.056 (0.60)
33.4 (2.0)
0.143, 0.168
127.30, 123.06
44.24–1.78
56,484 (1,608)
99.76 (96.0)
10.5 (4.2)
0.047 (0.55)
37.1 (2.1)
0.157, 0.185
127.73, 123.64
24.86–2.02
38,149 (1,269)
96.70 (99.0)
7.2 (6.9)
0.093 (0.70)
14.2 (2.1)
0.160, 0.198
Resolution (ꢁ)
Unique reflections
Completeness (%)
Redundancy
[a]
[b]
Rmerge
I/s (I)
Rwork , Rfree
[c]
[d]
[e]
Ramachandran statistics
Favored (%)
97.60
97.80
97.52
Allowed (%)
2.40
2.20
2.48
Outliers (%)
0
0
0
Number of protein, solvent and ligand
atoms
Average B factors (ꢁ )
3,238, 517, 52
3,211, 445, 65
3,218, 394, 53
2
Protein
Solvent
FMN
22.21
35.20
16.22
23.20
34.40
19.17
31.40
38.87
26.21
Ligands
33.30, 28.40,
34.90, 31.80, 38.0 (malo-
nate);
39.30, 26.20 (malonate);
27.01
23.10
2
1.70 (nicotinamide)
(p-chlorophenol)
[
[
[
[
[
a]
b]
c]
d]
e]
Values in parentheses denote data for the highest resolution bin (1.53–1.50 ꢁ, 1.81–1.78 ꢁ, 2.09–2.02 ꢁ).
Rmerge =S jIꢀ<I> j/S<I>.
Rwork =SjF
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(hkl)jꢀjF
A
T
N
T
E
N
N
ACHTUNGTRENNUNG
o
o
c
Rfree is calculated in the same manner as R
Statistics generated using MOLPROBITY.
using 10% of the reflection data not included during the refinement.
work
[18]
[
4a]
confirmed by a more systematic investigation. We latter is particularly critical for protein engineering
therefore sought old yellow enzyme homologs pos- studies of this enzyme, solving the X-ray crystal struc-
sessing residues other than Trp at the position corre- ture of OYE 2.6 has been a priority and the results
sponding to 116 of S. pastorianus OYE1. This prompt- are reported here. In addition, we also provide the re-
ed us to clone and overexpress the OYE 2.6 gene sults of four site-saturation mutagenesis libraries that
from the xylose-fermenting yeast Pichia stipitis CBS allow for meaningful comparisons with S. pastorianus
[8]
6
054. Based on sequence alignment and computer OYE1 and also point the way toward future ap-
modeling from the S. pastorianus OYE1 structure, Ile proaches to obtaining even more useful variants of
13 in OYE 2.6 was expected to occupy a similar OYE 2.6.
1
active site location as Trp 116. On this basis, we ex-
pected that the stereoselectivity of OYE 2.6 would be
analogous to that of the S. pastorianus OYE1 Trp 116 Results and Discussion
Ile mutant; however, the reality proved more com-
plex.
Native OYE 2.6 was overexpressed in E. coli and pu-
In general, OYE 2.6 has proven to be a superior rified by ammonium sulfate fractionation followed by
[
10]
alkene reductase with regard to stereoselectivity, con- affinity
and gel filtration chromatography. After
[
9]
version rate and stability. In some cases, its stereose- screening approximately 300 conditions, crystals were
lectivity was opposite that of S. pastorianus OYE1, observed when organic salts were the main precipi-
but this was not universally true. There are currently tants at near-neutral pH values. The most promising,
two impediments to the wider use of OYE 2.6 in obtained in 2.4M sodium malonate, pH 7.0, diffracted
chemical synthesis: a relatively limited knowledge of to a maximum resolution of 2.0 ꢁ using a Rigaku Cu
substrate and stereoselectivity characteristics and an rotating anode X-ray source. Unfortunately, these
unknown three-dimensional structure. Because the proved unsuitable for crystallographic studies as they
2
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

Structural and Catalytic Characterization of Pichia stipitis OYE 2.6
were not singular and yielded diffraction patterns that
could not be indexed accurately. Because attempts at
optimizing pH and salt concentrations proved unsuc-
cessful, a further screening of 96 additives coupled
with the original conditions (2.4M sodium malonate,
pH 7.0) was carried out. Buffer containing 1–3% 2-
propanol yielded crystals that diffracted to a maximum
usable resolution of 1.5 ꢁ using synchrotron source
radiation. The unit cell measured 127.1ꢀ123.4 ꢁ and
the crystals belonged to space group P6 22. The
3
asymmetric unit contained one molecule and a solvent
content of 63% with a Matthews coefficient of
3
[11]
3
.23 ꢁ /Dal.
The initial OYE 2.6 structure (with no added
[
12]
ligand) was solved by molecular replacement using
[
13]
S. pastorianus OYE1 as the search model (1OYA).
The final OYE 2.6 model was refined to 1.5 ꢁ resolu-
tion with final values of Rwork and Rfree of 0.143 and
0.168, respectively (Table 1). The enzyme exhibits the
eight-stranded a/b-barrel fold first seen in triosephos-
phate isomerase (TIM) and glycolate oxidase (GOX)
and belongs to the TIM phosphate binding family
[14]
(
Figure 1, A). The overall structure of OYE 2.6 is
similar to those of other old yellow enzyme homologs,
particularly OYE1 from S. pastorianus. These share
43% sequence identity and also have closely related
backbone positioning (aligning 325 a-carbons yielded
a root mean square deviation of 0.92 ꢁ; Figure 1, B).
The next most similar structures are Pseudomonas
[15]
putida morphinone reductase (MR, 1GWJ)
(291
Ca aligned, 1.26 ꢁ r.m.s.d.), Enterobacter cloacae pen-
taerythritol tetranitrate reductase (PETNR, 1H50)
Figure 1. Structure of P. stipitis OYE 2.6. A: Overall struc-
ture of the OYE 2.6 homodimer. The dashed line indicates
[
16]
(
232 Ca aligned, 1.38 ꢁ r.m.s.d.), followed by Bacillus the molecular two-fold axis, which is parallel to the crystal-
[17]
subtilis YqjM (1Z41)
(191 Ca aligned, 2.57 ꢁ lographic a,b plane. The enzyme exhibits a TIM barrel fold
and the backbone is colored by secondary structure motifs
r.m.s.d.). The most important structural difference be-
tween P. stipitis OYE 2.6 and S. pastorianus OYE1 is
found in loop L6 (Glu 287–Ala 302) (Figure 1, B).
Most of the key active site residues are conserved be-
tween OYE 2.6 and OYE1, suggesting that these can
be grouped into the same class of OYE homologs. S.
(
a-helices, cyan; b-sheets, magenta; loops, pink). Bound li-
gands (FMN and nicotinamide) are shown as sticks. The sol-
vent-accessible surface is shown for reference. B: Aligned
structures of P. stipitis OYE 2.6 (cyan) and S. pastorianus
OYE1 (orange). Enzyme backbones are depicted in ribbon
representation with regions of highest deviation shown as
cartoons for clarity. Numbers refer to the OYE 2.6 amino
[
19]
pastorianus OYE1 is dimeric, both in solution and
[
13]
in the crystal state. Gel filtration chromatography acid sequence. Loop 6 lies above the ribityl chain of the
indicated that P. stipitis OYE 2.6 is also a dimer in so- FMN cofactor and is outlined by a dashed circle.
lution (data not shown), and an analysis of subunit-
subunit contacts is also consistent with a dimer that
2
buries a total of 1520 ꢁ at the interface region. This other OYE family members. The well-defined and
[
20]
2
was further corroborated by PISA analysis,
which well-ordered cofactor (all B-factorsꢁ10 ꢁ ) is bound
predicted that OYE 2.6 would form a stable homodi- in the center of the enzymeꢃs parallel 8 strand b-
mer with a predicted DGdiss =5.0 kcal/mole). Interest- barrel (see Supporting Information). Its negatively
ingly, despite crystallizing in a different space group charged phosphate group interacts directly with Arg
(
P6 22 for OYE 2.6 versus P4 2 2 for S. pastorianus 347, a highly conserved residue among members of
3
3
1
OYE1), the relative positions of the two monomers the family. Oxygen O3’ appears to hydrogen bond
are nearly identical and, in both cases, the monomers with both water 31 and Arg 240, whose side chain
are related by a 2-fold crystallographic symmetry op- also interacts indirectly with O2’. The dimethylben-
erator.
zene moiety of FMN makes contacts with the aromat-
The constellation of amino acid side chains that in- ic residues Phe 373 and Tyr 374 and its isoalloxazine
teract with FMN in OYE 2.6 is highly conserved in ring is positioned to form direct hydrogen bonds with
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3
ÞÞ
These are not the final page numbers!

Yuri A. Pompeu et al.
FULL PAPERS
pastorianus OYE1 depended upon both hydrogen
bonding interactions with the phenolACTHUNGTRENNNUG
[
22c]
the FMN reduction potential.
nolic oxygen lies near the side chains of His 191 and
Asn 194; by contrast, two histidines occupy these po-
sitions in OYE 2.6 (His 188 and 191). Compared with
the corresponding values for S. pastorianus OYE1,
lmax values for both p-chlorophenol and p-hydroxy-
benzaldehyde interacting with OYE 2.6 are blue-shift-
ed by 59 and 27 nm, respectively. Relatively small ef-
fects on charge-transfer lmax values were observed
from changing phenolACHTUNGTRENNUNG
[
23]
ners in OYE1,
[24]
tentials led to more significant wavelength shifts.
Given the similarity in active site structures between
OYE 2.6 and OYE1, the differences in charge-transfer
lmax values are therefore more likely to reflect small
differences in flavin electronic properties. Using the
relationships between FMN E8’ and lmax values for p-
chlorophenol and p-hydroxybenzaldehyde established
[
22c]
for S. pastorianus OYE1,
we estimated that the E8’
value for OYE 2.6 is ca. ꢀ215 mV. This is similar to
[24]
that measured for wild-type OYE1 (ꢀ230 mV).
The initial OYE 2.6 crystals displayed electron den-
sity above the si face of the FMN that could be well
fit by malonate, which was present in the crystalliza-
tion buffer (Figure 3, A). In this complex, one carbox-
ylate oxygen was positioned to form hydrogen bonds
with the side chains of histidines 188 and 191, which
likely mimics their interactions with the carbonyl oxy-
gens of conjugated alkene substrates (vide infra). The
other carboxylate oxygen formed a hydrogen bond
with an ordered water molecule (WAT 301) that also
interacted with the side-chains of Tyr 78 and Thr 35.
While several well-ordered malonate molecules were
Figure 2. Phenol binding to OYE 2.6. A: p-Chlorophenol;
aliquots of p-chlorophenol were added to a 21.0 mM solution
of OYE 2.6 to give final ligand concentrations of 2, 4, 6, 8,
1
0, 12, 16, 24, 28, 32, 36, 52, 80 and 120 mM. UV-Vis spectra observed in the crystal, only this active site malonate
were recorded after each addition. Inset: A versus [p- was found in the higher energy planar conformation
chlorophenol]. B: p-Hydroxybenzaldehyde; aliquots of p- (all solvent-exposed malonates were observed in the
586
hydroxybenzaldehyde were added to a 26.6 mM solution of
OYE 2.6 to give final ligand concentrations of 4, 8, 12, 16,
expected bent form).
Crystals grown from a second preparation of OYE
2
0, 30 and 70 mM. UV-Vis spectra were recorded after each
2.6 also showed additional electron density in the
558
addition. Inset: A versus [p-hydroxybenzaldehyde].
active site region; however, the shape differed signifi-
cantly from that observed previously and was consis-
Gln 111, Thr 35, Arg 240 and the main chain nitrogen tent with free nicotinamide at an occupancy of ca.
[
25]
of Ala 68. Gln 111 is highly conserved in the OYE 95% (Figure 3, B and Table 1). This fortuitous com-
family and this side chain, along with that of Thr 35, plex likely mirrors that between OYE 2.6 and its nat-
is implicated in regulating the FMN reactivity in S. ural substrate, NADP(H). The amide carbonyl
[
21]
pastorianus OYE1.
oxygen is located within hydrogen bond distances of
We used spectral changes that accompany phenol the imidazole side chains of both His 188 and His 191
binding to OYE 2.6 to characterize its active site char- and the distance and angle between the nicotinamide
[
22]
acter and cofactor properties (Figure 2).
OYE 2.6 C4 and the FMN are consistent with the geometry of
[26]
bound both p-chlorophenol and p-hydroxybenzalde- hydride transfer (3.4 ꢁ and 948, respectively). Fur-
hyde tightly and these complexes were accompanied ther supporting the notion that this arrangement
by long wavelength charge transfer bands with lmax at mimics the productive NADPH/OYE 2.6 Michaelis
ca. 586 and 558 nm, respectively. Massey determined complex, the nicotinamide binding orientation within
that the positions of these charge transfer bands for S. OYE 2.6 is analogous to those observed for tetrahy-
4
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

Structural and Catalytic Characterization of Pichia stipitis OYE 2.6
dronicotinamide within the active sites of Thermo-
[
27]
AHCTUNGTNERNaGU naerobacter pseudethanolicus TOYE and morphi-
[28]
none reductase. Unfortunately, neither UV-Vis nor
fluorescence spectroscopy yielded usable spectral
changes for nicotinamide binding to OYE 2.6, so its
binding constant could not be measured by these
methods.
To provide a model for how potential substrates
might bind to the active site of OYE 2.6, the p-chloro-
phenol complex was prepared by soaking crystalline
OYE 2.6 in cryoprotectant buffer containing 2 mM in-
hibitor (Figure 3, C). The phenolic oxygen was posi-
tioned between the side chains of His 188 and 191
and a meta ring carbon was 3.5 ꢁ above the FMN N5
atom. The angle between this carbon and N5 and N10
of the FMN was 1058. These values are consistent
with those expected for efficient hydride transfer to
a cyclohexenone b-carbon.
After structural characterization of OYE 2.6 was
completed, we turned our attention to probing its
structure/function relationships with respect to alkene
reductions. We chose three representative Baylis–Hill-
man adducts 1–3 for catalytic studies of OYE 2.6
(
Figure 4) since they possess additional functionality
that makes them suitable for further useful chemical
transformations following asymmetric alkene reduc-
[
29]
tions.
While a number of catalytic hydrogenation
strategies for the asymmetric reduction of 3 has been
[
30]
reported (for recent examples, see ref. and referen-
ces cited therein), we are not aware of similar cata-
lysts being applied to cyclic enones 1 and 2. In addi-
tion, we have previously used the same series for
characterizing S. pastorianus OYE1 mutants, thereby
facilitating direct comparison of their substrate and
[
4a]
stereoselectivities.
We created four site-saturated mutagenesis libraries
of OYE 2.6 using a variation of the Reymond method
and a reduced genetic alphabet (NNK) that contained
32 codons but still encompassed all twenty amino
Figure 4. Baylis–Hillman adducts tested as substrates.
Figure 3. Ligand interactions with the active site of OYE
.6. A: Malonate complex; the simulated annealing compo-
site omit map contoured at the 1.5 s level is shown for
bound malonate and two ordered water molecules. Potential
hydrogen bonds with the side chains of histidines 188 and
vent molecule. Potential hydrogen bonds are indicated by
dashed lines. C: p-Chlorophenol complex; the simulated an-
nealing composite omit map contoured at the 0.5 s level is
shown for the ligands and an ordered water molecule. Par-
tial occupation by malonate and p-chlorophenol was ob-
served and modeled. Potential hydrogen bonds are indicated
by dashed lines.
2
191 are shown by dashed lines. B: Nicotinamide complex;
the simulated annealing composite omit map contoured at
the 1.2 s level is shown for the ligand and an ordered sol-
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
5
ÞÞ
These are not the final page numbers!

Yuri A. Pompeu et al.
FULL PAPERS
[31]
acids.
After extensive optimization, libraries rou- investigated here. As observed in the previous case,
tinely contained ꢂ16 amino acid replacements (usual- the catalytic activities of the mutant proteins were
ly 18 or 19). After verifying their diversity at the generally consistent across all three enone substrates.
pooled plasmid DNA stage, this mixture was used to This underscores the point that the catalytic assays
transform the E. coli expression host, individual reveal general characteristics of the enzymes and sug-
clones were seeded in 95 wells (the last contained gests that similar behavior will apply to other poten-
wild-type control) and all plasmids present in the tial substrates as well. The analogous position in S.
plate were sequenced. For every randomized amino pastorianus OYE1 (194) contains Asn. Interestingly,
acid position, a representative of each codon (when attempts to prepare the single Asn 194 His mutant of
[
23]
available) was arrayed on one third of a 96 well plate S. pastorianus OYE1 were unsuccessful. The analo-
to eliminate redundant screening efforts. In the case gous His in E. cloacae PETNR (position 184) also
of the Ile 113 library, simple site-directed mutagenesis showed limited tolerance for amino acid replacement,
methods were used to obtain the three missing although the Asn mutant retained 20% of the wild-
[
33,34]
codons so that a complete set of amino acid replace- type activity for 2-cyclohexenone.
Several
ments could be screened.
PETNR library members showed improved stereose-
[
35]
Enones 1–3 were reduced under whole-cell condi- lectivity for nitroalkenes.
tions in the presence of glucose for NADPH regener-
In both OYE 2.6 and S. pastorianus OYE1, the
ation by host metabolic pathways. OYE 2.6 was over- side-chain of a key Thr residue forms a hydrogen
expressed as a fusion protein with glutathione S-trans- bond with O4 of the bound FMN. For OYE1, loss of
ferase. Reaction conditions (cell mass and reaction this interaction (in the Thr 37 Ala mutant) slowed the
time) were chosen so that the wild-type enzyme af- reductive half-reaction by an order of magnitude but
[
21a]
forded ca. 99% conversion for substrates 1 and 2 and increased the oxidative half-reaction.
Other OYE
75% conversion for substrate 3 during the biotransfor- family members contain Thr or Cys at the analogous
mations. Reaction progress and product enantiopuri- position. Based on these precedents, we created and
ties were assessed by chiral-phase GC. Controls reac- examined a systematic replacement library for Thr 35
tions verified that racemization did not occur during in OYE 2.6 (Figure 5, A). For OYE 2.6, the Ser re-
the biotransformations (data not shown).
placement had essentially the same properties as the
Histidine 188 in OYE 2.6 showed little tolerance wild-type enzyme, consistent with the importance of
toward amino acid substitutions and only the wild- hydrogen bonding with the FMN cofactor. The Ala
type protein was highly active (see Supporting Infor- variant also possessed activity, albeit diminished. The
mation). The Ser, Gln and Asn variants reduced most surprising result was that several other residues
1
and 2 poorly but showed no detectable activity (Cys, Leu, Met, Gln and Val) could also functionally
toward 3. All variants retaining catalytic activity replace Thr 35, though with lower efficiencies. No
showed the same (S)-stereoselectivity as the wild-type change in stereoselectivity was found for any of the
enzyme. In S. pastorianus OYE1, the corresponding Thr 35 substitutions. The observed tolerance toward
position (191) is also occupied by His. This residue changes at this position bodes well for future work in
was changed to Asn by Massey; while the variant re- active site remodeling.
tained catalytic activity, the K value for 2-cyclohex-
Substitutions at the analogous position of other
M
[
23]
Ae none increased significantly.
C
H
T
U
N
G
T
R
E
N
N
U
N
G
The differential re- OYE family members have also been reported. Like
sponse toward analogous substitutions highlights the OYE 2.6, this position is occupied by Thr in wild-type
subtle differences in active site architecture, even E. cloacae PETNR (position 26). A site-saturation li-
among highly similar alkene reductases. Despite the brary of replacements gave very low average activity;
lack of success in identifying functional histidine re- however, both Ala and Ser substitutions were func-
[
33]
placements in this particular case, others have found tional. As part of a larger iterative saturation muta-
that site-saturation libraries can sometimes reveal un- genesis study of B. subtilis YqjM, Bougioukou et al.
[32]
expected results. Scrutton mutated the correspond- examined substitutions for Cys 26 (corresponding to
[
36]
ing His residue in E. cloacae PETNR (position 181) Thr 35 in OYE 2.6).
For one model enone sub-
and found that this library had very low average cata- strate, YqjM tolerated a number of substitutions at
lytic activity when tested against carbonyl-conjugated this position, although the impacts on conversion and
[33]
alkenes. On the other hand, several useful replace- stereoselectivity were modest. Interestingly, the effect
ments for E. cloacae PETNR His 181 were identified of these substitutions was increased significantly by
[34]
for nitroalkenes.
additional changes elsewhere in the active site, sup-
Position 191 in OYE 2.6 was slightly more tolerant porting the notion that this position in OYEs plays an
toward mutation, with both the native His and the important cooperative role in governing their proper-
Asn mutant proteins showing significant catalytic ac- ties.
tivity (see Supporting Information). No changes in
Our previous studies of S. pastorianus OYE1 high-
enantioselectivity were observed for the three enones lighted the key influence of Trp 116 in controlling re-
6
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

Structural and Catalytic Characterization of Pichia stipitis OYE 2.6
[
4a,7]
activity
and similar observations have been made
[
33]
for both E. cloacae PETNR
and B. subtilis
[
36]
YqjM. As noted above, the corresponding position
in native OYE 2.6 (113) is occupied by Ile. A number
of amino acids was tolerated by OYE 2.6 at position
113 (Figure 5, B). Small-to-medium hydrophobic resi-
dues (Ala, Leu, Met, Phe and Val) as well as polar,
uncharged amino acids (Asn, Cys, Gln, His, Ser and
Thr) could replace the native Ile with good to excel-
lent retention of catalytic activity. On the other hand,
charged amino acids, Pro and larger hydrophobic resi-
dues (Tyr and Trp) yielded inactive proteins toward
the Baylis–Hillman adducts examined here. One dif-
ference between the position 113 mutants and the li-
braries described above is that conversions differ sig-
nificantly between the two cyclic enones and the acy-
clic Roche ester precursor.
With regard to substrate conversion, the response
of OYE 2.6 toward amino acid substitutions at posi-
tion 113 closely mirrors that of a Trp 116 replacement
[
4a]
library in S. pastorianus OYE1 (Figure 5, C). Like
OYE 2.6, OYE1 prefers cyclic enones 1 and 2 over
acyclic alkene 3 and similar ranges of residues at posi-
tion 116 are acceptable substitutes for the native Trp.
In some cases, this similarity extends to changes in
stereoselectivity. For example, wild-type OYE1 (with
Trp at position 116) reduces 2 with predominantly
(
R)-selectivity (60% ee); by contrast, the Trp 116 Ile
[
4a]
mutant afforded 91% ee (S)-product. For OYE 2.6,
wild-type (with Ile at position 113) reduced 2 with
high (S)-selectivity (>90% ee) while the Ile 113 Trp
mutant provided 11% ee favoring the (R)-enantiomer.
One important difference between the behavior of
OYE 2.6 and S. pastorianus OYE1 mutants involves
Roche ester precursor 3. Wild-type OYE 2.6 and all
of the Ile 113 mutants that accept this substrate cata-
lyze reductions with essentially complete (S)-selectivi-
ty, although some with relatively poor conversion. By
contrast, mutating Trp 116 in S. pastorianus OYE1
provides a range of stereoselectivities for this sub-
strate, ranging from >98% ee (R) (wild-type) to
[
4a]
>
98% ee (S) (Trp 116 Gln). These differences high-
light the distributive nature of the substrate binding
pocket within alkene reductases. While single amino
substitutions can be sufficient to alter stereoselectivity
in some favorable cases, multiple substitutions are
more often required. Comparing the structures of
OYE 2.6 and S. pastorianus OYE1 in light of these
Figure 5. Site-saturation mutagenesis results for key active
site residues. A: Thr 35 of OYE 2.6. OYE 2.6 variants with
all possible replacements for Thr 35 (except Trp and Lys)
were screened for reductions of Baylis–Hillman adducts 1, 2
and 3 using whole-cell assays. Arrows indicate data for the
wild-type enzyme and amino acids shown in grey were
absent from the library. B: Ile 113 of OYE 2.6; variants with
all possible replacements for Ile 113 were screened for re-
ductions of Baylis–Hillman adducts 1, 2 and 3 using whole-
cell assays. Arrows indicate data for the wild-type enzyme.
C: Trp 116 of S. pastorianus OYE1; variants with all possible
replacements for Trp 116 were screened for reductions of
Baylis–Hillman adducts 1, 2 and 3 using whole-cell assays.
Arrows indicate data for the wild-type enzyme.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
7
ÞÞ
These are not the final page numbers!

Yuri A. Pompeu et al.
FULL PAPERS
mutagenesis data provides a much better roadmap for proteins also contain an analogous ordered water
future protein engineering studies.
molecule. Interestingly, this water was only observed
near the FMN in OYE 2.6. In the case of OYE1, MR,
PETNR and OPR1, a tryptophan residue fills this
volume and eliminates the possibility of an ordered
water molecule at this location. Both YqjM and
Conclusions
A major motivation for determining the crystal struc- TOYE possess Ala in this location that provides suffi-
ture of OYE 2.6 was to understand its stereoselectiv- cient space; however, the two residues that form hy-
ity, particularly why it is generally opposite to that of drogen bonds with the ordered water in OYE 2.6
wild-type S. pastorianus OYE1 for Baylis–Hillman ad- (Thr 35 and Tyr 78) either cannot form hydrogen
ducts 1–3 (but identical to that of the OYE1 W116I bonds (Ile) or have a suitable residue (Cys), but in
mutant). In two of our three OYE 2.6 structures, an unfavorable orientations. Taken together, this may ex-
ordered water molecule hydrogen bonds to the side- plain the absence of this critical ordered water mole-
chains of Thr 35 and Tyr 78 (Figure 3, A and B). cule in the other proteins.
When the FMN moieties of the OYE 2.6/malonate
The positions of the active site Thr side-chains (po-
complex and the S. pastorianus W116I OYE1/enone 2 sitions 35 and 37 in OYE 2.6 and S. pastorianus
complex were overlaid, the hydroxy group of 2 was in OYE1, respectively) are essentially indistinguishable.
the same location as these ordered water molecules in The same is true for the Ile side-chain (Ile 113 in
OYE 2.6 (Figure 6). Our previous results have argued OYE 2.6 and the W116I side-chain in S. pastorianus
that this protein-substrate interaction in S. pastorianus OYE1), with one important difference. In the OYE1
OYE1 plays a key role in determining which face of mutant, the Ile side-chain was observed in two confor-
the alkene p system will accept hydride from reduced mations, only one of which allows productive sub-
[4a]
FMN. It is therefore reasonable to suggest that an strate binding. Likewise, two binding orientations of
analogous hydrogen bond is also formed between enone 2 were observed in the complex with W116I S.
OYE 2.6 and the hydroxymethyl groups of 1–3 and pastorianus OYE1, one potentially relevant to cataly-
[
4a]
this provides a structural rationale for the observed sis and the other clearly incorrect. By contrast, the
stereoselectivity. We carried out structural alignments corresponding Ile side-chain in OYE 2.6 (Ile 113) was
[
13]
[4a]
of OYE 2.6 with wild-type and the W116I mutant
S. pastorianus OYE1, morphinone reductase (MR),
pentaerythritol tetranitrate reductase (PETNR),
only found in a single conformation, which matched
the productive arrangement in the W116I S. pasto-
AHCTUNGTRENNUGNr ianus OYE1 mutant (Figure 6). This singular confor-
[
[
15]
16]
[17]
YqjM,
Thermoanaerobacter pseudethanolicus E39 mation of Ile 113 may be one reason that OYE 2.6 is
TOYE (3 KRZ) and tomato oxophytodienoate re- a more efficient catalyst when compared to the
ductase OPR1 (1ICS) to determine whether these W116I mutant of S. pastorianus OYE1.
Despite these similarities, there are also key differ-
[27]
[
37]
ences between OYE 2.6 and S. pastorianus OYE1 –
even the OYE1 W116I mutant. For example, when re-
ducing (S)-carvone, both wild-type S. pastorianus
OYE1 and OYE 2.6 share the same stereoselectivity,
[7]
which is opposite that of the OYE1 W116I mutant.
This not only points out the special nature of Baylis–
Hillman substrate binding to alkene reductase active
sites, but also underscores the fact OYE 2.6 is not
simply the same as the W116I mutant of S. pastoria-
nus OYE1. Other active site residues of OYE 2.6
must also influence the orientation of substrate bind-
ing. We are currently preparing and evaluating addi-
Figure 6. Comparison of OYE 2.6 and S. pastorianus OYE1 tional OYE 2.6 mutants and their complexes with po-
complexes. The structures of native OYE 2.6 with bound tential substrates in order to understand these factors
malonate (cyan, 3TJL) and the structure of the S. pastoria-
more completely.
nus W116I OYE1 mutant with enone 2 present in the pro-
ductive binding orientation (green, 3RND) were overlaid
using all FMN atoms (yellow carbons). Key side-chains and
an ordered water molecule are shown (OYE 2.6 numbering
is shown with the corresponding residues for S. pastorianus
Experimental Section
OYE1 in parentheses). For clarity, only the OYE 2.6 atoms
are shown for the side-chain of Thr 35 and the FMN cofac-
General
tor since the corresponding atoms of S. pastorianus OYE1
are found in identical positions.
Restriction endonucleases, Phusion Hot Start II High-Fideli-
ty DNA Polymerase and T4 DNA ligase were purchased
8
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

Structural and Catalytic Characterization of Pichia stipitis OYE 2.6
from New England Biolabs. Primers were obtained from In-
dialysis buffer supplemented with 10 mM sodium dithionite
for 6 h. In the final dialysis step (8 h), sodium dithionite was
omitted from the buffer, yielding a bright yellow dialysate.
The crude protein sample was loaded onto an N-(4-hydroxy-
tegrated DNA Technologies. Authentic (R)-Roche ester
2
D
2
{
measured [a] : ꢀ15.9 (c 5.59, CHCl )} was obtained from
3
Sigma. Crystallography screening kits (Classics Suite II/
[10]
AmSO and Additives kit) and were purchased from Qiagen
benzoyl)aminohexyl agarose affinity column
previously
4
and Hampton Research, respectively. All other reagents
were obtained from commercial suppliers and used as re-
ceived. Plasmids were purified on small scales by EconoSpin
columns (Epoch Life Sciences) and on large scales using
CsCl density gradient ultracentrifugation. DNA sequenc-
ing was carried out by the University of Florida ICBR using
capillary fluorescence methods using either standard proto-
cols (single samples) or rolling circle amplification (96 well
plates). LB medium contained 5 g/L Bacto-Yeast Extract,
equilibrated with 100 mM Tris-Cl, 50 mM (NH ) SO ,
pH 8.0. The column matrix became dark brown, indicating
successful OYE 2.6 binding. After loading, the column was
4
2
4
washed with the starting buffer at
a flow rate of
[38]
ꢀ1
0.5 mLmin until the A280 value of the eluant stabilized at
ca. 0.2. OYE 2.6 was eluted by washing the column with de-
oxygenated starting buffer supplemented with 4 mM sodium
dithionite. The protein fractions with the most intense
yellow color (after subsequent air oxidation) were pooled
and concentrated by ultrafiltration to approximately
1
0 g/L Bacto-Tryptone and 10 g/L NaCl. ZYP-5052 auto-in-
ducing media contained 10 g/L tryptone, 5 g/L yeast extract,
mM MgSO , 25 mM (NH ) SO , 50 mM KH PO , 50 mM
ꢀ1
10 mgmL . Final purification was achieved by gel filtration
chromatography on a Superdex 200 column (Pharmacia)
equilibrated with 30 mM Tris-Cl, 30 mM NaCl pH 7.5. The
eluted protein was concentrated by ultrafiltration to ca.
1
4
4
2
4
2
4
Na HPO , 5 g/L glycerol, 0.5 g/L anhydrous glucose and
2
4
[39]
2
g/L a-lactose monohydrate.
ꢀ1
280
[40]
Chiral-phase GC analyses were carried out with a 30 mꢀ
40 mgmL (using a calculated e value of 55,810) prior
0.25 mm b-Dex 225 column (Supelco) with He as the carrier
to crystallization studies.
gas and detection by FID. For analysis of substrates 1 and 2
and their reduction products, the temperature program in-
Crystallogenesis
ꢀ
1
volved 1408C for 10 min followed by a 208Cmin increase
to 2008C, which was maintained for 3 min. Under these con-
ditions, the (S)- and (R)-reduction products from 1 eluted at
Crystals were grown using the sitting drop vapor diffusion
method from 2.4M sodium malonate, pH 7.0 buffer supple-
mented with 1–3% 2-propanol. After growth, crystals were
soaked in harvesting buffer (3.0M sodium malonate, 10% v/
v glycerol, pH 7.0) and flash frozen in liquid nitrogen for
10.1 min and 10.6 min, respectively, and the (R)- and (S)-re-
duction products from 2 eluted at 10.4 and 11.4 min, respec-
tively. Starting materials 1 and 2 eluted at 13.0 and 13.1 min,
respectively. For analysis of enone 3 and its reduction prod-
ucts, the temperature program involved 1008C for 12 min
data collection. The crystals belonged to space group P6 22.
3
To prepare the p-chlorophenol complex, crystals were brief-
ly soaked in harvesting buffer supplemented with 2 mM p-
chlorophenol. Increasing soaking times increased mosaicity
and decreased the resolution of the crystals. The best results
ꢀ
1
followed by a 208Cmin increase to 1808C, which was
maintained for 5 min. Under these conditions, the (S)- and
(
R)-reduction products eluted at 10.8 and 11.4 min, respec-
(
diffraction to 2.02 ꢁ) resulted from soaking for 5 seconds.
tively, and the starting material was observed at 12.1 min.
Representative chromatograms of starting materials and rac-
emic products are provided in the Supporting Information.
Structure Solution
Reflection data were processed using the HKL3000 program
[
41]
suite.
Phases were obtained by molecular replacement
Expression and Purification of Native P. stipitis OYE
[12]
using the AutoMR utility of PHENIX. using a modifica-
tion of S. pastorianus OYE1 (PDB code 1OYA) as the
search model. All ligands, water molecules and non-identical
side-chains were removed prior to molecular replacement.
2.6
E. coli BL21 (DE3) harboring plasmid pBS3 (a derivative of
pET 26b containing the OYE 2.6 coding region flanked by
NdeI and XhoI sites) was grown at 378C in LB medium sup-
plemented with 50 mg/mL kanamycin. IPTG was added to
a final concentration of 400 mM when the culture had
reached an OD600 value of 0.6 and shaking was continued at
One solution was found in space group P6 22. Moreover,
3
the initially calculated 2F -F and F -F maps showed elec-
o
c
o
c
tron density patterns that could be easily identified as FMN,
further validating the molecular replacement solution. The
initial model consisted of 385 amino acids (most devoid of
side chains) and an Rfree value of 0.33. After the initial simu-
lated annealing refinement followed by refinement with in-
dividual xyz parameters and B-factors, the Rfree dropped to
0.24. Inspection of the newly-phased electron density maps
allowed placement of the missing residues and showed that
some regions of the protein backbone required rebuilding.
Refining the more complete model improved the phases
and electron density maps clearly showed both ordered sol-
vent molecules and amino acid side chains. Iterative cycles
of model-building and refinement were continued to yield
the final protein structures. Model building was carried out
3
08C for an additional 8–10 h. Cells were harvested by cen-
trifugation at 5,000ꢀg, resuspended in 100 mM Tris-Cl,
0 mM PMSF, pH 8.0 at a concentration of 1 g/mL and lysed
1
by a French pressure cell at 12,000 psi. All purification steps
were carried out at 48C. Debris was removed by centrifuga-
tion (18,000ꢀg for 45–60 min). Solid ammonium sulfate was
added to the supernatant to reach 78% saturation, then in-
soluble material was removed by centrifugation (18,000ꢀg
for 30 min). The pellet was resuspended in a minimal
volume of 100 mM Tris-Cl, 50 mM (NH ) SO , pH 8.0, then
4
2
4
protamine sulfate was added to a final concentration of
mg/mL and the resulting insoluble material was removed
1
[42]
by centrifugation (15,000ꢀg for 20 min). The dark yellow
supernatant was dialyzed against 100 mM Tris-Cl, 50 mM
with COOT and refinement was performed by PHENIX.
After the protein and cofactor modeling was completed, ad-
ditional electron density lying above the FMN was apparent
(
NH ) SO , pH 8.0 for 8 h. This was replaced by additional
4 2 4
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
9
ÞÞ
These are not the final page numbers!

Yuri A. Pompeu et al.
FULL PAPERS
Table 2. Primers used for library creation.
Position
Sequences
35 Forward
35 Reverse
113 Forward
113 Reverse
188 Forward
188 Reverse
191 Forward
191 Reverse
5’-ATCCACCAACTNNKAGATTTAGAGCTTTAGAAGACCACAC-3’
5’-GCTCTAAATCTMNNAGTTGGTGGATAGACAATCTTGGTTT-3’
5’-CAACCCAGTTGNNKTTTTTGGGAAGGGTTGCAGATCCAGC-3’
5’-CTTCCCAAAAAMNNCAACTGGGTTGAAACGAAAGAACCG-3’
5’-ATATTGAACTCNNKGCTGCTCATGGCTACCTTTTAGATCA-3’
5’-CCATGAGCAGCMNNGAGTTCAATATAGTCGAAACCAGCAT-3’
5’-TCCATGCTGCTNNKGGCTACCTTTTAGATCAATTTTTGCA-3’
5’-AAAAGGTAGCCMNNAGCAGCATGGAGTTCAATATAGTCGA-3’
in both the 2mF -F and F -F maps. Either malonate alone
LB medium, then pelleted by brief centrifugation. Pooled
plasmid DNA was isolated by alkaline lysis, purified by spin
columns and sequenced by automatic fluorescence methods
to identify samples with the highest degree of degeneracy at
the desired positions.
o
c
o
c
(
3TJL) or nicotinamide (3UPW) completely accounted for
this additional active site electron density. In the case of the
p-chlorophenol soaked structure, the final electron density
maps suggested only partial occupancy of this ligand (ap-
proximately 60%), consistent with the very short ligand
soaking time. Additional electron density beyond that of p-
chlorophenol was modeled as a malonate molecule in the
same orientation as observed in the native structure. The de-
posited structure (4DF2) has both ligands in the model. All
protein structure figures were created using PyMOL (Schrç-
dinger, LLC).
The pooled degenerate plasmid sample (1 ng) was used to
transform 40 mL of electrocompetent BL21 Gold (DE3)
cells. After recovery for 1 h at 378C in 600 mL of SOC
medium, cells were spread on LB agar plates containing
ꢀ1
200 mgmL ampicillin and incubated overnight at 378C.
Ninety five individual transformants (plus a wild-type con-
trol) were used to seed a 2 mL deepwell plate and all 96
plasmids were prepared by rolling circle amplification and
sequenced. Based on the sequencing data, clones represent-
ing each of the 32 possible codons (when available) were
Phenol Binding Studies
ꢀ
1
Assays were performed at 208C in 100 mM KP , pH 7.0 in
i
used to seed 600 mL of LB media containing 200 mgmL
1
a total volume of 1 mL. Purified OYE 2.6 was added to
a final concentration of ca. 20 mM. The actual protein con-
centration present during each experiment was calculated
ampicillin arrayed on = of a deepwell plate (2 mL total
3
volume per well) to yield library master plates. The number
of codons and amino acids obtained from each library are
listed in Table 3.
4
60
460
[22a]
from the initial value of A using e =10,600.
of the appropriate phenol (p-chlorophenol or p-hydroxy-
ACHTUNGTRENNUNGb enzaldehyde; 4.0 mM stocks in EtOH) were added to the
Aliquots
When sequence data from the Ile 113 library were ana-
lyzed, OYE 2.6 genes encoding the Asp, Lys and Phe var-
iants were missing. These were prepared individually by
a modification of the method described above. For these
amplifications, primers encoded only a single amino acid at
position 113 (instead of NNK degeneracy), 20 ng of tem-
plate (pBS2) were used and one extension temperature
(728C) was employed; other conditions were identical. Am-
plicons were purified, digested with DpnI and aliquots
protein solution and the mixtures were stirred briefly. Spec-
tra were recorded at ca. 1 min intervals until no further
change was observed (typically ꢃ3 min). This process was
continued until apparent saturation was reached.
Library Creation and Screening
Site saturation mutagenesis libraries were prepared by
(
5 mL) were used to transform E. coli JM109 cells (75 mL)
[31]
a modification of the methods reported by Zheng et al.
by electroporation. Plasmid DNA was purified from ran-
domly chosen colonies (spin columns) and sequenced to
identify the missing OYE 2.6 variants. The appropriate plas-
mids were used to transform E. coli BL21 Gold (DE3) cells
Each PCR mixture (total volume 100 mL) containing 5ꢀ
Phusion HF Buffer (20 mL), template pBS2 (1 ng), forward
and reverse NNK degenerate primers (0.5 mM each), dNTPs
(
200 mM each), and Phusion Hot Start II High-Fidelity
DNA Polymerase (1 U) was subjected to an initial denatura-
tion step of 988C (30 s) followed by 25 cycles of 988C (10 s)
and 62–728C (4 min) followed by a final incubation at 728C
Table 3. Library completeness data.
Position Number of Se-
Unique
codons
Unique amino
acids
(
7 min). Primer sequences are shown in Table 2. Amplicons
were purified by DNA spin columns, digested with DpnI at
78C (10 U for 4 h followed by an additional 10 U for 4 h),
[a]
quences
3
Thr 35 79
29
26
31
28
18
17
then purified again by DNA spin columns. Aliquots (5 mL)
were used to transform Electo-Ten Blue (Stratagene) electo-
competent cells (75 mL) by electroporation. SOC medium
was added (600 mL), then samples were incubated for 1 h at
[b]
Ile 113 84
His 188 92
His 191 91
19
16
[
a]
3
78C prior to selection on LB media plates supplemented
This indicates the subset of the 96 sequenced clones that
provided usable data.
The three missing amino acids were subsequently added
to this library individually.
ꢀ
1
with 200 mgmL ampicillin. After overnight incubation, all
colonies from transformations yielding ꢂ 300 c.f.u. from
three combined plates were suspended in a small volume of
[b]
10
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

Structural and Catalytic Characterization of Pichia stipitis OYE 2.6
and the resulting strains were used to fill in the missing
spaces in the Ile 113 library master plate.
[8] a) Also known as Scheffersomyces stipitis; b) T. W. Jef-
fries, I. V. Grigoriev, J. Grimwood, J. M. Laplaza, A.
Aerts, A. Salamov, J. Schmutz, E. Lindquist, P. Dehal,
H. Shapiro, Y.-S. Jin, V. Passoth, P. M. Richardson,
Nature Biotech. 2007, 25, 319–326.
Library screening plates were prepared from library
master plates, either freshly grown in LB medium supple-
ꢀ
1
mented with 200 mgmL ampicillin or from frozen stocks.
In the former case, the library master plate was shaken at
[9] D. J. Bougioukou, A. Z. Walton, J. D. Stewart, Chem.
200 rpm and 378C. After 6 h, the cultures were visibly
Commun. 2010, 46, 8558–8560.
turbid and a 20-mL aliquot from each well was used to inoc-
ulate wells in a duplicate deepwell plate containing 600 mL
of ZYP-5052 auto-inducing medium supplemented with
[10] A. S. Abramovitz, V. Massey, J. Biol. Chem. 1976, 251,
5321–5326.
[11] B. W. Matthews, J. Mol. Biol. 1968, 33, 491–497.
[12] P. D. Adams, P. V. Afonine, G. Bunkoczi, V. B. Chen,
I. W. Davis, N. Echols, J. J. Headd, L. W. Hung, G. J.
Kapral, R. W. Grosse-Kunstleve, A. J. McCoy, N. W.
Moriarty, R. Oeffner, R. J. Read, D. C. Richardson,
J. S. Richardson, T. C. Terwilliger, P. H. Zwart, Acta
Crystallogr. Sect. D 2010, 66, 213–221.
ꢀ1
200 mgmL
ampicillin. Alternatively, the auto-induction
plates were inoculated directly from frozen library master
plates. After seeding, library screening plates were then
mounted in a growth apparatus designed to facilitate maxi-
mal oxygen transfer rates (based on the design of Duetz and
[43]
Witholt ) and shaken at 300 rpm and 378C. After 16–18 h,
cells were harvested by centrifuging at 3,000 rpm for 30 min
and the supernatant was removed by aspiration. Cell pellets
[
[
13] K. M. Fox, P. A. Karplus, Structure 1994, 2, 1089–1105.
14] N. Nagano, C. A. Orengo, J. M. Thornton, J. Mol. Biol.
2002, 321, 741–765.
were resuspended in 300 mL of 50 mM KP , 100 mM glucose,
i
10 mM substrate, pH 7.0. The plate was shaken at 200 rpm
[
[
[
[
15] T. Barna, H. L. Messiha, C. Petosa, N. C. Bruce, N. S.
at room temperature. After 6 h, wells were individually ex-
tracted with 500 mL of EtOAc prior to chiral-phase GC anal-
ysis.
Scrutton, P. C. F. Moody, J. Biol. Chem. 2002, 277,
3
0976–30983.
16] T. M. Barna, H. Khan, N. C. Bruce, I. Barsukov, N. S.
Scrutton, P. C. F. Moody, J. Mol. Biol. 2001, 310, 433–
4
47.
17] K. Kitzing, T. B. Fitzpatrick, C. Wilken, J. Sawa, G. P.
Acknowledgements
Bourenkov, P. Macheroux, T. Clausen, J. Biol. Chem.
2
005, 280, 27904–27913.
Financial support by the National Science Foundation (CHE-
18] V. B. Chen, W. B. Arendall, J. J. Headd, D. A. Keedy,
R. M. Immormino, G. J. Kapral, L. W. Murray, J. S. Ri-
chardson, D. C. Richardson, Acta Crystallogr. Sect. D
0615776 and CHE-1111791) and the U.S. Army Advanced
Civilian Schooling Program is gratefully acknowledged. We
also thank Professor Steven Bruner for his generous guidance
in crystallographic studies, Professor Robert McKenna and
Dr. Arthur Robbins (U.F. Center for Structural Biology) for
diffractometer access and many helpful discussions and Dr.
Annie Hꢀroux [Macromolecular Crystallography Research
Resource (PXRR), National Synchotron Light Source] for
crystallographic data collection.
2
010, 66, 12–21.
[
[
[
19] K. Saito, D. J. Thiele, M. Davio, O. Lockridge, V.
Massey, J. Biol. Chem. 1991, 266, 20720–20724.
20] E. Krissinel, K. Henrick, J. Mol. Biol. 2007, 372, 774–
7
97.
21] a) D. Xu, R. M. Kohli, V. Massey, Proc. Natl. Acad. Sci.
USA 1999, 96, 3556–3561; b) B. J. Brown, J. W. Hyun, S.
Duvvuri, P. A. Karplus, V. Massey, J. Biol. Chem. 2002,
2
77, 2138–2145.
22] a) R. G. Matthews, V. Massey, J. Biol. Chem. 1969, 244,
779–1786; b) R. G. Matthews, V. Massey, C. C. Swee-
[
References
1
ley, J. Biol. Chem. 1975, 250, 9294–9298; c) A. S. Abra-
movitz, V. Massey, J. Biol. Chem. 1976, 251, 5327–5336.
23] B. J. Brown, Z. Deng, P. A. Karplus, V. Massey, J. Biol.
Chem. 1998, 273, 32753–32762.
[
1] a) R. E. Williams, N. C. Bruce, Microbiology 2002, 148,
1
607–1614; b) R. Stuermer, B. Hauer, M. Hall, K.
[
[
[
Faber, Curr. Opin. Chem. Biol. 2007, 11, 203–213;
c) H. S. Toogood, J. M. Gardiner, N. S. Scrutton, Chem-
CatChem 2010, 2, 892–914; d) M. Hall, A. S. Bommar-
ius, Chem. Rev. 2011, 111, 4088–4110.
24] R. C. Stewart, V. Massey, J. Biol. Chem. 1985, 260,
3
639–3647.
25] While this electron density could also be fit well by
benzoate, we favor nicotinamide since it was present in
the bacterial growth medium while benzoate was not.
[26] M. W. Fraaije, A. Mattevi, Trends Biochem. Sci. 2000,
25, 126–132.
[27] B. V. Adalbjçrnsson, H. Toogood, A. Fryszkowska,
C. R. Pudney, T. A. Jowitt, D. Leys, N. S. Scrutton,
ChemBioChem 2010, 11, 197–207.
[28] C. R. Pudney, S. Hay, J. Y. Pang, C. Costello, D. Leys,
M. J. Sutcliffe, N. S. Scrutton, J. Am. Chem. Soc. 2007,
129, 13949–13956.
[
[
2] Formerly known as Saccharomyces carlsbergensis.
3] A. D. N. Vaz, S. Chakraborty, V. Massey, Biochemistry
1
995, 34, 4246–4256.
[
4] a) A. Z. Walton, W. C. Conerly, Y. Pompeu, B. Sullivan,
J. D. Stewart, ACS Catalysis 2011, 1, 989–993; b) E.
Brenna, F. G. Gatti, A. Manfredi, D. Monti, F. Parmeg-
giani, Eur. J. Org. Chem. 2011, 4015–4022.
5] P. F. Mugford, U. G. Wagner, Y. Jiang, K. Faber, R. J.
Kazlauskas, Angew. Chem. 2008, 120, 8912–8923;
Angew. Chem. Int. Ed. 2008, 47, 8782–8793.
[
[
[
6] D. J. Bougioukou, Ph.D. thesis, University of Florida
(Gainesville, FL, USA), 2006.
[29] a) C. Stueckler, C. K. Winkler, M. Bonnekessel, K.
Faber, Adv. Synth. Catal. 2010, 352, 2663–2666; b) A.
Chen, J. Xu, W. Chiang, C. L. L. Chai, Tetrahedron
7] S. K. Padhi, D. J. Bougioukou, J. D. Stewart, J. Am.
Chem. Soc. 2009, 131, 3271–3280.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
11
ÞÞ
These are not the final page numbers!

Yuri A. Pompeu et al.
FULL PAPERS
2
010, 66, 1489–1495; c) N. Mase, A. Inoue, M. Nishio,
K. Takabe, Bioorg. Med. Chem. Lett. 2009, 19, 3955–
958.
[36] D. J. Bougioukou, S. Kille, A. Taglieber, M. T. Reetz,
Adv. Synth. Catal. 2009, 351, 3287–3305.
[37] C. Breithaupt, J. Straßner, U. Breitinger, R. Huber, P.
Macheroux, A. Schaller, T. Clausen, Structure 2001, 9,
419–429.
[38] J. Sambrook, E. F. Fritsch, T. Maniatis, Molecular Clon-
ing. A Laboratory Manual, 2nd edn., Cold Spring
Harbor Press, Cold Spring Harbor, 1989.
3
[
30] a) C. Pautigny, S. Jeulin, T. Ayad, A. Zhang, J.-P.
GenÞt, V. Ratovelomanana-Vidal, Adv. Synth. Catal.
2
008, 350, 2525–2532; b) J. Wassenaar, M. Kuil, M.
Lutz, A. L. Spek, J. N. H. Reek, Chem. Eur. J. 2010, 16,
509–6517; c) P. Etayo, J. L. NfflÇez-Rico, A. Vidal-
6
Ferran, Organometallics 2011, 30, 6718–6725.
31] L. Zheng, U. Baumann, J.-L. Reymond, Nucl. Acids
Res. 2004, 32, e115.
[
39] F. W. Studier, Protein Expression Purif. 2005, 41, 207–
34.
[
[
[
2
[
40] S. Duvaud, M. R. Wilkins, R. D. Appel, A. Bairoch, in:
The Proteomics Handbook, (Ed.: J. M. Walker),
Humana Press, Totowa, NJ, 2005, pp 571–607.
32] A. Yep, G. L. Kenyon, M. J. McLeish, Proc. Natl. Acad.
Sci. USA 2008, 105, 5733–5738.
33] M. Hulley, H. S. Toogood, A. Fryszkowska, D. Mansell,
G. M. Stephens, J. M. Gardiner, N. S. Scrutton, Chem-
BioChem 2010, 11, 2433–2447.
34] H. Toogood, A. Fryszkowska, M. Hulley, M. Sakuma,
D. Mansell, G. M. Stephens, J. M. Gardiner, N. S. Scrut-
ton, ChemBioChem 2011, 12, 738–749.
35] A. Fryszkowska, H. Toogood, M. Sakuma, G. M. Ste-
phens, J. M. Gardiner, N. S. Scrutton, Catal. Sci. Tech-
nol. 2011, 1, 948–957.
[
[
[
41] Z. Otwinowski, W. Minor, Meth. Enzymol. 1997, 276,
3
07–326.
42] P. Emsley, B. Lohkamp, W. G. Scott, K. Cowtan, Acta
Crystallogr. Sect. D 2010, 66, 486–501.
43] W. A. Duetz, B. Witholt, Biochem. Eng. J. 2001, 7, 113–
[
[
1
15.
12
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
Structural and Catalytic Characterization of Pichia stipitis
OYE 2.6, a Useful Biocatalyst for Asymmetric Alkene
Reductions
13
Adv. Synth. Catal. 2012, 354, 1 – 13
Yuri A. Pompeu, Bradford Sullivan, Adam Z. Walton,
Jon D. Stewart*
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
13
ÞÞ
These are not the final page numbers!